https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69275
Bug ID: 69275 Summary: ICE compiling rs6000/float-128.c Product: gcc Version: 6.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: bergner at gcc dot gnu.org Target Milestone: --- With revision 232346 (ie, Mike Meissner's last IEEE128 patch), we're now seeing the following ICE on powerpc64le-linux: /home/bergner/gcc/build/gcc-fsf-mainline-cpu-supports-base/./gcc/xgcc -B/home/bergner/gcc/build/gcc-fsf-mainline-cpu-supports-base/./gcc/ -B/home/bergner/gcc/install/gcc-fsf-mainline-cpu-supports-base/powerpc64le-linux/bin/ -B/home/bergner/gcc/install/gcc-fsf-mainline-cpu-supports-base/powerpc64le-linux/lib/ -isystem /home/bergner/gcc/install/gcc-fsf-mainline-cpu-supports-base/powerpc64le-linux/include -isystem /home/bergner/gcc/install/gcc-fsf-mainline-cpu-supports-base/powerpc64le-linux/sys-include -g -O2 -O2 -g -O2 -DIN_GCC -W -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual -Wno-format -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition -isystem ./include -fPIC -mlong-double-128 -mno-minimal-toc -g -DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector -fPIC -mlong-double-128 -mno-minimal-toc -I. -I. -I../.././gcc -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/. -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/../gcc -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/../include -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/../libdecnumber/dpd -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/../libdecnumber -DHAVE_CC_TLS -Wno-type-limits -mvsx -mfloat128 -mno-float128-hardware -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/soft-fp -I/home/bergner/gcc/gcc-fsf-mainline-base/libgcc/config/rs6000 -o float128-ifunc.o -MT float128-ifunc.o -MD -MP -MF float128-ifunc.dep -c /home/bergner/gcc/gcc-fsf-mainline-base/libgcc/config/rs6000/float128-ifunc.c -fvisibility=hidden -DHIDE_EXPORTS /home/bergner/gcc/gcc-fsf-mainline-base/libgcc/config/rs6000/float128-ifunc.c:282:8: error: ifunc is not supported on this target TFtype __addkf3 (TFtype, TFtype) ^~~~~~~~ /home/bergner/gcc/gcc-fsf-mainline-base/libgcc/config/rs6000/float128-ifunc.c:285:8: error: ifunc is not supported on this target TFtype __subkf3 (TFtype, TFtype) ^~~~~~~~ Mike, I assume you bootstrapped using a POWER9 enabled (ie, new) binutils when you tested the patch. If I use an older binutils (eg, 2.24 like on our genoa system), then we get the error above.